Nvidia AI ML Performance Engineer 

United States, Washington 

709344553

What You’ll Be Doing:

• Develop high-fidelity performance models to prototype emerging algorithmic techniques in Generative AI to drive model-hardware co-design.

• Design targeted optimizations for inference deployment to maximize Pareto frontier of Accuracy, Throughput and Interactivity.

• Quantify performance benefit of targeted optimizations to prioritize features and guide future software and hardware roadmap.

• Model end-to-end performance impact of emerging GenAI workflows - such as Agentic Pipelines, Inference-time compute scaling, etc. - to guide datacenter design and optimization.

• This position requires you to keep up with the latest DL research and collaborate with diverse teams, including DL researchers, hardware architects, and software engineers.

What we need to see:

• A minimum qualification of a Master's degree (or equivalent experience) in Computer Science, Electrical Engineering or related fields.

• Strong background in computer architecture, roofline modeling, queuing theory and statistical performance analysis techniques.

• Solid understanding of LLM internals (attention mechanisms, FFN structures), model parallelism and inference serving techniques.

• 3+ years of hands-on experience in system evaluation of AI/ML workloads or performance analysis, modeling and optimizations for AI.

• Proficiency in Python (and optionally C++) for simulator design and data analysis.

• Growth mindset and pragmatic “measure, iterate, deliver” approach.

Ways to Stand Out from the Crowd:

• Comfortable defining metrics, designing experiments and visualizing large performance datasets to identify resource bottlenecks.

• Proven track record of working in cross-functional teams, spanning algorithms, software and hardware architecture.

• Ability to distill complex analyses into clear recommendations for both technical and non-technical stakeholders.

• Experience with GPU computing (CUDA)

• Experience with deep learning frameworks like PyTorch, TRT-LLM, VLLM, SGLang